Complex, vast, and fast-changing habitats: How AI can help

14 Mar 2024 | 8 min read

Complex, vast, and fast-changing habitats: How AI can help

The biodiversity plan, now a cornerstone of any development, outlines how biodiversity net gain will be achieved. All net gain plans require a biodiversity metric calculation, which establishes the units a habitat contains before development takes place – measuring units lost and determining what is needed to achieve a 10% BNG. The formula takes different factors into account, including the habitat’s size, condition, location, and type. Accurate habitat maps are essential to BNG calculations and to biodiversity plans as, if the baseline is wrong, so will be everything else that follows.

When mapping habitats for BNG, there are various factors to consider and, as we have outlined in “The importance of integrating collaborative AI technology with ecological expertise to meet biodiversity net gain targets,” AI can help ecologists solve the problems inherent in the traditional habitat mapping process, saving time, and reducing the risk of costly inaccuracies.

In this blog, we will explore how our tools at AiDash can assist ecologists when dealing with complex, vast and fast-changing habitats in particular.

With an estimated 150,000 BNG project applications needing to be submitted each year, achieving the quality and speed dictated by the 2021 Environment Act it is not feasible with the ecologist resource available. Only through collaboration with AI and satellite technology can ecologists be supported and enabled to ensure the successful implementation of BNG.

Restoring and managing habitats

Restoring and managing habitats is a multifaceted undertaking that relies heavily on accurate mapping and strategic planning. The process involves ecologists conducting BNG assessments by physically visiting sites to record intricate habitat details, a task that is both challenging and time-consuming. Given the shortage of ecologists in the UK, there is a pressing need for a more efficient approach. Addressing this gap, AiDash has developed a BNG AI portal designed to streamline and enhance the process.

Prior to setting foot on the site, a combination of satellite imagery and AI technology can be employed to conduct a preliminary assessment producing a habitat map. This also involves integrating various data layers, including government records and soil data, to generate insights into the likely habitats present. Having this information prepopulated and stored in one place not only saves time and effort but also proves invaluable for ecologists classifying habitats on the ground. With knowledge into the historical states of these habitats, high-value features can be flagged for the ecologist to verify on foot, preventing oversights and enhancing the accuracy and efficiency of the mapping process.

Furthermore, satellite imagery proves instrumental in quantifying certain habitats. For instance, it can detect small areas of scrub and bracken emerging on grasslands, which may evolve overtime into distinct habitats. These can eventually dominate a significant portion of the site. Determining the coverage of these habitats is crucial to the BNG metric and must be as accurate as possible. Satellite technology aids in detecting these subtle shifts, contributing to more effective habitat management and conservation efforts.

Vast habitats

The expansive nature of certain habitats poses a challenge when it comes to accurate scaling and delineation – a task often impractical for on-the-ground assessments. Satellites provide a comprehensive view of large habitats, offering detailed insights into their extents and conditions. This is particularly beneficial for certain wet habitats, such as saltmarshes and bogs which are often less accessible on foot.

Additionally, mapping the boundaries and composition of habitats is more precise and visually accessible from an aerial perspective. For instance, it can be difficult to reliably estimate the mix of conifers and broadleaf trees in woodland based upon impressions gained during the woodland walk that is used to assess the condition of the habitats as not all of the habitat will be visited and visibility into the wood is obscured by trees and shrubs. AI and satellite imagery can accurately map the woodlands, distinguishing between different woodland types for on-site ecologists to confirm, streamlining the overall assessment.

The integration of AI and satellite technology enables quicker and larger-scale estimations, aligning seamlessly with the new BNG framework. AiDash’s BNG AI portal can generate habitat maps for numerous sites before physical exploration, not only saving time but also ensuring the project is compliant with the new regulations at every step.

Fast changing habitats

Some habitats in the UK undergo rapid changes influenced by both natural processes like plant succession and human-induced alterations resulting from land management practices. When mapping sites for BNG, it’s important to be aware of these dynamic habitats that can experience shifts in type, size or condition relatively quickly.

Free satellite tools typically display the most useful photo of the habitat for the needs of the general public, not necessarily the most recent one and time stamps are often absent. Variability in the age of the image, the season, and its provenance, make forming an objective, reliable, consistent analysis of land cover difficult. Images from a year ago may already be inaccurate. Relying on free imagery reduces the quality of the work and the control of an analyser over their own process.

AiDash’s BNG AI portal simplifies this process by utilizing recent, high-resolution satellite images of habitats. This guarantees the availability of essential information, like time stamps, for making well-informed decisions. The process can be repeated at future intervals, and all the data is retained in the platform, providing a full life cycle documentation approach to the projectTraditional mapping methods may struggle to keep pace with these changes, whereas satellite data can track shifts more easily. This is crucial for demonstrating that habitats undergoing enhancement or changes are progressing as intended, and those not intended to change are identified early on.

Collaboration for a single point of truth

Habitats are intricate and difficult to classify, making successful implementation of BNG possible only through collaboration between ecologists and AI and satellite technology. The newly established regulations are rigorous, aiming to ensure a much-needed positive impact.

While on-the-ground validation remains essential, technology plays a pivotal role in easing the burden on ecologists, particularly in challenging scenarios like defining boundaries around large bodies of water. Accuracy in maps is crucial for BNG, yet data misalignment can occur in large-scale projects tackled by ecologists short of time and resources. Satellite technology addresses this challenge by breaking down sites into smaller, non-overlapping portions and mapping surrounding infrastructure. The precision required becomes apparent as minor inaccuracies surface during the conversion of surveys into detailed BNG maps.

AiDash’s BNG AI portal ensures that every square meter is accounted for. Serving as the foundational backbone for a project, the portal not only harmonizes data to present a single point of truth but also delivers a detailed, precise, and compliant habitat map from which to build an effective, compliant biodiversity plan.

Related Posts

What's the difference? Navigating distinctions between preliminary...
tr{border: solid 1px #000;} td{border: solid 1px #000;} th{border: solid...
Read more
BNG applications ask for detail beyond earlier...
As developers move forward to comply with biodiversity net gain...
Read more
Barratt Developments’ approach to embracing BNG mandates
Biodiversity net gain (BNG) requirements have landed. Developers, local planning...
Read more